Award-winning artist Pras Michel is allegedly “grasping at straws” by asserting that his leading counsel provided inferior representation in a previous criminal trial by utilizing artificial intelligence (AI), testified Alon Israely, one of Michel’s legal professionals. Michel’s defense team had used a generative AI program from EyeLevel.AI to supplement its legal research. However, the team members testified that they did not exclusively rely on the AI’s responses.
Known in the world of music and recently, in political influencing, Michel was found guilty of ten criminal counts in April last year, which included a covert lobbying campaign to halt an investigation of the Malaysian business magnate, Jho Low.
There’s another twist in the plot as his new legal representation from ArentFoxSchiff is pushing for a fresh trial based on allegations that his chief lawyer in the former trial, David Kenner, delivered deficient and ineffective assistance. That assistance entails the alleged usage of an experimental AI program to draft his closing arguments. Michel further argues that a contempt charge pursued by prosecutors against Kenner for purportedly leaking grand jury material resulted in a conflict of interest for the counsel during the trial.
On the contrary, the US Justice Department claims that Michel’s accusations of receiving subpar assistance do not stand, citing the “overwhelming” evidence against him. The US District Judge presiding over the case, Colleen Kollar-Kotelly, has considered evidence in Michel’s motion for a new trial. Israely, a member of Kenner’s team, testified that information from EyeLevel.AI constituted only a few minutes of Kenner’s roughly three-hour closing argument.
Interestingly, Kenner admitted in his testimony that his use of generative AI for the closing argument caused him to wrongly attribute lyrics from a Puff Daddy song to the Fugees, Michel’s group. “I messed up,” Kenner said. He firmly added that if the court considers this as ineffective counsel, it should allow a motion for a new trial.
More details about the case can be found at “USA v. Michel“.
This unique case provides a view into the complexities of AI use in legal proceedings, particularly in high-stakes white-collar cases such as Michel’s. While AI and machine learning have the potential to aid in legal research and counsel, this situation highlights how its misuse can lead to contentious outcomes.
This information is based on the article on Bloomberg Law.